Low-power Wireless Sensor Networking in Interference Environments

Abstract

The demand for commercial deployment of large-scale wireless sensor networks (WSNs) has rapidly been increasing over the past decade. However, conventional WSN technologies may not be feasible for commercial deployment of large-scale WSNs because of their technical flaws, including limited network scalability, susceptibility to co-channel interference and large signaling overhead. In practice, low-power WSNs seriously suffer from interference generated by coexisting radio systems such as IEEE 802.11 wireless local area networks (WLANs). This interference problem seriously hampers commercial deployment of low-power WSNs. Few commercial WSN chips can provide secure and reliable networking performance in practical operation environments. In this dissertation, we consider performance improvement of low-power WSNs in the presence of co-channel interference. We first investigate the effect of co-channel interference on the transmission of low-power WSN signal, and then design a low-power WSN transceiver that can provide stable performance even in the presence of severe co-channel interference, while providing the backward compatibility with IEEE 802.15.4. We also consider the network connectivity in the presence of co-channel interference. The connectivity of low-power WSNs can be improved by transmitting synchronization signal and making channel hand-off in a channel-aware manner. A beacon signal for the network synchronization is repeatedly transmitted in consideration of channel condition and signaling overhead. Moreover, when the channel is severely interfered, all devices in a cluster network make communications by means of temporary channel hopping and then seamlessly make channel hand-off to the best one among the temporary hopping channels. The performance improvement is verified by computer simulation and experiment using IEEE 802.15.4 motes in real operation environments. Finally, we consider the signal transmission in the presence of co-channel interference. The throughput performance of low-power WSN transceivers can be improved by adjusting the transmission rate and the payload size according to the interference condition. We estimate the probability of transmission failure and the data throughput, and then determine the payload size to maximize the throughput performance. It is shown that the transmission time maximizing the normalized throughput is little affected by the transmission rate, but rather by the interference condition. The transmission rate and the transmission time can independently be adjusted in response to the change of channel and interference condition, respectively. The performance improvement is verified by computer simulation.